My favorites | Sign in
Project Logo
Project Home
  
Downloads
  
Wiki
  
Issues
  
Source
    
Summary | Updates | People
Details: Show all Hide all

Last 30 days

  • Dec 15, 2009
    r70 (Added a test case for initial values with whitespaces.) committed by fady.mou...@gmail.com   -   Added a test case for initial values with whitespaces.
    Added a test case for initial values with whitespaces.
  • Dec 14, 2009
    r69 (Reduced lexer complexity by using a case insensitive charact...) committed by fady.mou...@gmail.com   -   Reduced lexer complexity by using a case insensitive character stream and left factoring keywords.
    Reduced lexer complexity by using a case insensitive character stream and left factoring keywords.

Earlier this year

  • Oct 19, 2009
    issue 19 (Provide a standalone executable jar as an alternative to ant...) Labels changed by fady.mou...@gmail.com   -  
    Labels: Milestone-Release0.0.5
    Labels: Milestone-Release0.0.5
  • Oct 19, 2009
    issue 19 (Provide a standalone executable jar as an alternative to ant...) Status changed by fady.mou...@gmail.com   -  
    Status: Fixed
    Status: Fixed
  • Oct 19, 2009
    ANTClassLoadingIssues (Issues running ANT script.) Wiki page edited by fady.mou...@gmail.com   -   Revision r68 Typos
    Revision r68 Typos
  • Oct 19, 2009
    ANTClassLoadingIssues (Issues running ANT script.) Wiki page edited by fady.mou...@gmail.com   -   Revision r67 Added more cases.
    Revision r67 Added more cases.
  • Oct 19, 2009
    legstar-pli2cob-0.0.5.zip (Beta release 0.0.5) file uploaded by fady.mou...@gmail.com   -  
    Labels: Featured Type-Archive OpSys-All
    Labels: Featured Type-Archive OpSys-All
  • Oct 19, 2009
    ANTClassLoadingIssues (Issues running ANT script.) Wiki page edited by fady.mou...@gmail.com   -   Revision r66 Typos
    Revision r66 Typos
  • Oct 19, 2009
    ANTClassLoadingIssues (Issues running ANT script.) Wiki page edited by fady.mou...@gmail.com   -   Revision r65 Edited wiki page through web user interface.
    Revision r65 Edited wiki page through web user interface.
  • Oct 19, 2009
    issue 19 (Provide a standalone executable jar as an alternative to ant...) reported by fady.mou...@gmail.com   -   So far, besides using the pli2cob API directly from java, the utility was available as an ant task only. Ant raises a number of issues because ANTLR has been very popular and there are serious chances that some older version of ANTLR or StringTemplates are in ANT_HOME/lib (see http://code.google.com/p/legstar- pli2cob/wiki/ANTClassLoadingIssues). We need an alternative that will always work and has minimal prerequisites (ideally only a JRE 1.5).
    So far, besides using the pli2cob API directly from java, the utility was available as an ant task only. Ant raises a number of issues because ANTLR has been very popular and there are serious chances that some older version of ANTLR or StringTemplates are in ANT_HOME/lib (see http://code.google.com/p/legstar- pli2cob/wiki/ANTClassLoadingIssues). We need an alternative that will always work and has minimal prerequisites (ideally only a JRE 1.5).
  • Oct 19, 2009
    ANTClassLoadingIssues (Issues running ANT script.) Wiki page added by fady.mou...@gmail.com   -   Revision r64 Started documentating ANT classpath issues.
    Revision r64 Started documentating ANT classpath issues.
  • Oct 19, 2009
    r63 ([maven-release-plugin] prepare for next development iteratio...) committed by fady.mou...@gmail.com   -   [maven-release-plugin] prepare for next development iteration
    [maven-release-plugin] prepare for next development iteration
  • Oct 19, 2009
    r62 ([maven-release-plugin] copy for tag legstar-pli2cob-0.0.5) committed by fady.mou...@gmail.com   -   [maven-release-plugin] copy for tag legstar-pli2cob-0.0.5
    [maven-release-plugin] copy for tag legstar-pli2cob-0.0.5
  • Oct 19, 2009
    r61 ([maven-release-plugin] prepare release legstar-pli2cob-0.0.5) committed by fady.mou...@gmail.com   -   [maven-release-plugin] prepare release legstar-pli2cob-0.0.5
    [maven-release-plugin] prepare release legstar-pli2cob-0.0.5
  • Oct 19, 2009
    r60 (New standalone executable jar becomes the preferred way to r...) committed by fady.mou...@gmail.com   -   New standalone executable jar becomes the preferred way to run the utility. This is because ant scripts are hitting a lot of problems with conflicting antlr/stringtemplates/log4j jars. Also continued changing PLI by PL/I in texts and convert to translate.
    New standalone executable jar becomes the preferred way to run the utility. This is because ant scripts are hitting a lot of problems with conflicting antlr/stringtemplates/log4j jars. Also continued changing PLI by PL/I in texts and convert to translate.
  • Oct 18, 2009
    r59 (Documentation update. Change PLI into the more official PL/I...) committed by fady.mou...@gmail.com   -   Documentation update. Change PLI into the more official PL/I and the verb "convert" to "translate" for consistency. Decided to use "translate" for language to language transformations while "convert" is used for data format to data format transformation.
    Documentation update. Change PLI into the more official PL/I and the verb "convert" to "translate" for consistency. Decided to use "translate" for language to language transformations while "convert" is used for data format to data format transformation.
  • Oct 16, 2009
    legstar-pli2cob-0.0.4.zip (Beta release 0.0.3) file uploaded by fady.mou...@gmail.com   -  
    Labels: Type-Archive Featured OpSys-All
    Labels: Type-Archive Featured OpSys-All
  • Oct 16, 2009
    issue 18 (COBOL long literal values should be split over multiple line...) changed by fady.mou...@gmail.com   -   Value literals are now properly split.
    Status: Fixed
    Labels: Milestone-Release0.0.4
    Value literals are now properly split.
    Status: Fixed
    Labels: Milestone-Release0.0.4
  • Oct 16, 2009
    issue 17 (Warnings produced by ANTLR should use commons logging rather...) changed by fady.mou...@gmail.com   -   Lexer and Parser generated by ANTLR are considered abstract and we implement inheriting Lexers and Parsers. This allow us to override error handling and use commons-logging rather than System.out. This technique still allow us to use ANTLRWorks to test the grammars. For the tree grammar though, there was a lot of code we needed to put somewhere so we have an abstract class that contains all the additional code including logging.
    Status: Fixed
    Labels: Milestone-Release0.0.4
    Lexer and Parser generated by ANTLR are considered abstract and we implement inheriting Lexers and Parsers. This allow us to override error handling and use commons-logging rather than System.out. This technique still allow us to use ANTLRWorks to test the grammars. For the tree grammar though, there was a lot of code we needed to put somewhere so we have an abstract class that contains all the additional code including logging.
    Status: Fixed
    Labels: Milestone-Release0.0.4
  • Oct 16, 2009
    issue 16 (Non declare PL/I statements should be silently ignored) changed by fady.mou...@gmail.com   -   All non declare statements are now ignored rather than causing the parser to stop. You might still get warnings from the lexer though if it encounters characters it does not recognize. Normally it should recover properly and ignore these characters.
    Status: Fixed
    Labels: Milestone-Release0.0.4
    All non declare statements are now ignored rather than causing the parser to stop. You might still get warnings from the lexer though if it encounters characters it does not recognize. Normally it should recover properly and ignore these characters.
    Status: Fixed
    Labels: Milestone-Release0.0.4
  • Oct 16, 2009
    issue 9 (Data item names should not be COBOL reserved words) changed by fady.mou...@gmail.com   -   Also added checking for name collisions. This might happen for long PL/I names because they get truncated to 30 characters (COBOL limitations) therefore conflicts are possible. In case of collision, pli2cob changes the last characters by numerics and checks for conflict again until it finds a unique COBOL name.
    Status: Fixed
    Labels: Milestone-Release0.0.4
    Also added checking for name collisions. This might happen for long PL/I names because they get truncated to 30 characters (COBOL limitations) therefore conflicts are possible. In case of collision, pli2cob changes the last characters by numerics and checks for conflict again until it finds a unique COBOL name.
    Status: Fixed
    Labels: Milestone-Release0.0.4
  • Oct 16, 2009
    r58 ([maven-release-plugin] prepare for next development iteratio...) committed by fady.mou...@gmail.com   -   [maven-release-plugin] prepare for next development iteration
    [maven-release-plugin] prepare for next development iteration
  • Oct 16, 2009
    r57 ([maven-release-plugin] copy for tag legstar-pli2cob-0.0.4) committed by fady.mou...@gmail.com   -   [maven-release-plugin] copy for tag legstar-pli2cob-0.0.4
    [maven-release-plugin] copy for tag legstar-pli2cob-0.0.4
  • Oct 16, 2009
    r56 ([maven-release-plugin] prepare release legstar-pli2cob-0.0.4) committed by fady.mou...@gmail.com   -   [maven-release-plugin] prepare release legstar-pli2cob-0.0.4
    [maven-release-plugin] prepare release legstar-pli2cob-0.0.4
  • Oct 16, 2009
    r55 (Important refactoring based on a more intensive use of ANTLR...) committed by fady.mou...@gmail.com   -   Important refactoring based on a more intensive use of ANTLR grammars. First the data items hierarchy is not created by the parser grammar directly. Previously, the grammar was creating a flat tree and then a custom class ASTNormalizer was recreating the hierarchy. ASTNormalizer is gone. There is still a need to enhance the AST produced by the parser grammar. A new tree walker, PLIStructureEnhancer now does that based on a tree grammar. Finally, the COBOL emitting code is now done by another tree grammar, PLIStructureCobolEmitter in association with a group of StringTemplates. ASTToCobol is gone. While I was at it, I also fixed: Issue 9 (COBOL names should not be reserved words), there is also a collision detection between COBOL names. Issue 16 , non declare PL/I statements are silently ignored Issue 17 , all code now uses commons logging Issue 18 , long COBOL values should be split on multiple lines
    Important refactoring based on a more intensive use of ANTLR grammars. First the data items hierarchy is not created by the parser grammar directly. Previously, the grammar was creating a flat tree and then a custom class ASTNormalizer was recreating the hierarchy. ASTNormalizer is gone. There is still a need to enhance the AST produced by the parser grammar. A new tree walker, PLIStructureEnhancer now does that based on a tree grammar. Finally, the COBOL emitting code is now done by another tree grammar, PLIStructureCobolEmitter in association with a group of StringTemplates. ASTToCobol is gone. While I was at it, I also fixed: Issue 9 (COBOL names should not be reserved words), there is also a collision detection between COBOL names. Issue 16 , non declare PL/I statements are silently ignored Issue 17 , all code now uses commons logging Issue 18 , long COBOL values should be split on multiple lines
  • Oct 16, 2009
    issue 18 (COBOL long literal values should be split over multiple line...) reported by fady.mou...@gmail.com   -   If a COBOL value, resulting form a PL/I INITIAL clause, is too long to fit on a line, it should be split over multiple line using COBOL continuation standards.
    If a COBOL value, resulting form a PL/I INITIAL clause, is too long to fit on a line, it should be split over multiple line using COBOL continuation standards.
  • Oct 06, 2009
    r54 (Adressing issue 16 and issue 17. PLI source code can now inc...) committed by fady.mou...@gmail.com   -   Adressing issue 16 and issue 17 . PLI source code can now include non declare PLI statements, they will be ignored and error messages and warnings produced directly by ANTLR are now forwarded to commons logging rather than simply using System.out.
    Adressing issue 16 and issue 17 . PLI source code can now include non declare PLI statements, they will be ignored and error messages and warnings produced directly by ANTLR are now forwarded to commons logging rather than simply using System.out.
  • Oct 06, 2009
    issue 17 (Warnings produced by ANTLR should use commons logging rather...) reported by fady.mou...@gmail.com   -   Reporting is currently mixed: the legstar messages are logged via apache commons logging while ANTLR errors/warning are simply routed to System.out. It would be better to have all messages routed to commons logging so that users can use either LOG4J or JRE logging.
    Reporting is currently mixed: the legstar messages are logged via apache commons logging while ANTLR errors/warning are simply routed to System.out. It would be better to have all messages routed to commons logging so that users can use either LOG4J or JRE logging.
  • Oct 04, 2009
    issue 16 (Non declare PL/I statements should be silently ignored) reported by fady.mou...@gmail.com   -   A PL/I source fragment might contain other PL/I statements than declares. Right now, the parser fails if it encounters such a non declare statement. The workaround is to edit the fragment and manually remove all non declare statements. It would be be nicer to simply have the parser ignore them.
    A PL/I source fragment might contain other PL/I statements than declares. Right now, the parser fails if it encounters such a non declare statement. The workaround is to edit the fragment and manually remove all non declare statements. It would be be nicer to simply have the parser ignore them.
  • Sep 29, 2009
    PLIStructureReference (PL/I Structure Abstract Syntax Tree.) Wiki page edited by fady.mou...@gmail.com   -   Revision r53 Linking to graph...
    Revision r53 Linking to graph...
  • Sep 29, 2009
    PLIStructureReference (PL/I Structure Abstract Syntax Tree.) Wiki page edited by fady.mou...@gmail.com   -   Revision r52 Linking to graph...
    Revision r52 Linking to graph...
  • Sep 29, 2009
    PLIStructureReference (PL/I Structure Abstract Syntax Tree.) Wiki page edited by fady.mou...@gmail.com   -   Revision r51 Linking to graph...
    Revision r51 Linking to graph...
  • Sep 29, 2009
    PLIStructureReference (PL/I Structure Abstract Syntax Tree.) Wiki page edited by fady.mou...@gmail.com   -   Revision r50 Linking to graph...
    Revision r50 Linking to graph...
  • Sep 29, 2009
    PLIStructureReference (PL/I Structure Abstract Syntax Tree.) Wiki page added by fady.mou...@gmail.com   -   Revision r49 Adding an initial AST reference graph.
    Revision r49 Adding an initial AST reference graph.
  • Sep 29, 2009
    issue 13 (Some attributes such as VARYINGZ are wrongly forced to follo...) changed by fady.mou...@gmail.com   -   Strings and arithmetic attributes can now appear in any order. The parser should be more tolerant then in release 0.0.2.
    Status: Fixed
    Labels: Milestone-Release0.0.3
    Strings and arithmetic attributes can now appear in any order. The parser should be more tolerant then in release 0.0.2.
    Status: Fixed
    Labels: Milestone-Release0.0.3
  • Sep 29, 2009
    issue 7 (Union attribute is not supported) changed by fady.mou...@gmail.com   -   UNIONS are now translated to COBOL REDEFINES. A number of important restrictions are discussed here http://code.google.com/p/legstar- pli2cob/wiki/PLIToCOBOLUnionVsRedefines.
    Status: Fixed
    Labels: Milestone-Release0.0.3
    UNIONS are now translated to COBOL REDEFINES. A number of important restrictions are discussed here http://code.google.com/p/legstar- pli2cob/wiki/PLIToCOBOLUnionVsRedefines.
    Status: Fixed
    Labels: Milestone-Release0.0.3
  • Sep 29, 2009
    legstar-pli2cob-0.0.3.zip (Alpha release 0.0.3) file uploaded by fady.mou...@gmail.com   -  
    Labels: Featured Type-Archive OpSys-All
    Labels: Featured Type-Archive OpSys-All
  • Sep 29, 2009
    r48 ([maven-release-plugin] prepare for next development iteratio...) committed by fady.mou...@gmail.com   -   [maven-release-plugin] prepare for next development iteration
    [maven-release-plugin] prepare for next development iteration
  • Sep 29, 2009
    r47 ([maven-release-plugin] copy for tag legstar-pli2cob-0.0.3) committed by fady.mou...@gmail.com   -   [maven-release-plugin] copy for tag legstar-pli2cob-0.0.3
    [maven-release-plugin] copy for tag legstar-pli2cob-0.0.3
  • Sep 29, 2009
    r46 ([maven-release-plugin] prepare release legstar-pli2cob-0.0.3) committed by fady.mou...@gmail.com   -   [maven-release-plugin] prepare release legstar-pli2cob-0.0.3
    [maven-release-plugin] prepare release legstar-pli2cob-0.0.3
  • Sep 29, 2009
    r45 (Fixing issues reported by findbugs and PMD.) committed by fady.mou...@gmail.com   -   Fixing issues reported by findbugs and PMD.
    Fixing issues reported by findbugs and PMD.
  • Sep 29, 2009
    r44 (Fixing issue 7. Union attribute is now supported and transla...) committed by fady.mou...@gmail.com   -   Fixing issue 7 . Union attribute is now supported and translates into COBOL REDEFINES. There are a number of difficulties which are described here: http://code.google.com/p/legstar- pli2cob/wiki/PLIToCOBOLUnionVsRedefines.
    Fixing issue 7 . Union attribute is now supported and translates into COBOL REDEFINES. There are a number of difficulties which are described here: http://code.google.com/p/legstar- pli2cob/wiki/PLIToCOBOLUnionVsRedefines.
  • Sep 29, 2009
    issue 15 (Typed structures and unions should be supported by the parse...) reported by fady.mou...@gmail.com   -   Right now only declared structures and unions are supported. In PL/I, structures and unions can be defined and then used as generic types. The parser should be able to deal with DEFINE STRUCTURE keywords.
    Right now only declared structures and unions are supported. In PL/I, structures and unions can be defined and then used as generic types. The parser should be able to deal with DEFINE STRUCTURE keywords.
  • Sep 29, 2009
    issue 14 (PL/I ORDINAL should translate to COBOL level 88) reported by fady.mou...@gmail.com   -   In theory, PL/I ORDINALS, should translate to COBOL level 88. There are a number of difficulties though: - ORDINALS are defined separately from the members using them. This means the ordinal definition is not part of the structure we parse. - The PL/I ordinal member names do not have as many restrictions as their COBOL level 88 counterparts - The fixed binary data that backs the ORDINAL has a precision that may be set or defaults depending on the range covered - Like fixed binary, ordinals can occupy a single byte of storage.
    In theory, PL/I ORDINALS, should translate to COBOL level 88. There are a number of difficulties though: - ORDINALS are defined separately from the members using them. This means the ordinal definition is not part of the structure we parse. - The PL/I ordinal member names do not have as many restrictions as their COBOL level 88 counterparts - The fixed binary data that backs the ORDINAL has a precision that may be set or defaults depending on the range covered - Like fixed binary, ordinals can occupy a single byte of storage.
  • Sep 29, 2009
    PLIToCOBOLUnionVsRedefines (Mapping PL/I UNION to COBOL REDEFINES.) Wiki page edited by fady.mou...@gmail.com   -   Revision r43 Revisited the samples after union support implementation.
    Revision r43 Revisited the samples after union support implementation.
  • Sep 28, 2009
    issue 13 (Some attributes such as VARYINGZ are wrongly forced to follo...) reported by fady.mou...@gmail.com   -   PL/I statements such as "dcl 1 A VARYINGZ CHAR(1);" are valid but not supported by our parser right now. Similarly, the PRECISION keyword cannot appear before BINARY while it should be supported.
    PL/I statements such as "dcl 1 A VARYINGZ CHAR(1);" are valid but not supported by our parser right now. Similarly, the PRECISION keyword cannot appear before BINARY while it should be supported.
  • Sep 28, 2009
    r42 (Refactored the Abstract Syntax Tree. The VARYING attributes ...) committed by fady.mou...@gmail.com   -   Refactored the Abstract Syntax Tree. The VARYING attributes are now sub attributes to STRING. Similarly all arithmetic attributes have been grouped under a new ARITHMETIC node.
    Refactored the Abstract Syntax Tree. The VARYING attributes are now sub attributes to STRING. Similarly all arithmetic attributes have been grouped under a new ARITHMETIC node.
  • Sep 23, 2009
    PLIToCOBOLUnionVsRedefines (Mapping PL/I UNION to COBOL REDEFINES.) Wiki page edited by fady.mou...@gmail.com   -   Revision r41 Fixed typos
    Revision r41 Fixed typos
  • Sep 23, 2009
    PLIToCOBOLUnionVsRedefines (Mapping PL/I UNION to COBOL REDEFINES.) Wiki page added by fady.mou...@gmail.com   -   Revision r40 Added a wiki on union versus redefines.
    Revision r40 Added a wiki on union versus redefines.
  • Sep 04, 2009
    PLIToCOBOLAlignmentIssues (Aligned/Unaligned data items handling.) Wiki page edited by fady.mou...@gmail.com   -   Revision r39 last typo.
    Revision r39 last typo.
 
Hosted by Google Code